BowFest 2026: what a field full of archers told me that no test suite could

Part of the "From Vision to Production" closing arc.

For eighteen months this series was about how the app got built — the method, the team, the quality system, the economics. BowFest was the first time none of that mattered. The app stopped being a thing I talked about building and became a thing people picked up and used.

A quick fact to keep the story straight: BowFest wasn't the launch. BowSmith has been live on the App Store since early July — three weeks before the festival. So this was the app's first public outing, not its debut. BowFest 2026 ran July 23–26 at Mont du Lac Resort in Superior, Wisconsin — billed as the world's largest outdoor archery and music festival, with 3D courses, night shoots, novelty shoots and live music.

Let me start with the honest number, because there isn't one. Downloads and signups didn't move in any way I can point to yet. If this were a launch post I'd bury that line. It isn't — so here's what actually happened, which turned out to be far more useful than a spike would have been.

What actually pulled people in

The interest didn't come from a pitch at a stand. It came from the session summaries I posted after my own shoots — and specifically from the analytics behind them. BowSmith takes your sessions and results and builds setup recommendations for your next session and for competitions, and that was the thing people leaned in on. Not the feature list. The idea that the app watches how you actually shoot and tells you what to change.

What they asked for that isn't there yet

The single most requested thing, over and over, was bowhunting. People wanted the hunting-side features, and right now they aren't in the app. I heard it clearly enough that it's changed the roadmap (more on that below). Worth saying plainly: the most valuable thing a field full of your actual users can do is tell you what's missing, and this weekend the answer was unambiguous.

What broke — the honest ledger

The habit that made the "what failed" post in this series land is the same one I owe this post. Something did break, and it was consistent enough to be a real finding: a number of people hit the same rough edge in nested navigation — doubled buttons for controlling forms, screens stacked in a way that got confusing.

The interesting part is where it showed up. It surfaced almost entirely in Tinkering sessions; Practice sessions worked fine. That's a useful signal — it points the fix at a specific area rather than a vague "navigation is off." It's now scoped as a cross-app navigation audit and fix, landing in 1.2.0.

What changes because of it

A recap is just a diary unless something moves. Three things did:

• 1.2.0 — the navigation fix. The nested-form/stack issue gets a proper cross-app audit, not a patch.
• Bowhunting joins the backlog. The most-requested gap at the festival is now a committed line item.
• Hardware moved from idle thought to real ideation. The conversations I didn't expect were about sensors — an Apple Watch companion, and even bow-mounted sensors on the riser (a Mathews-style sense). Still ideation, not a promise. But it started in a field in Wisconsin, not at a whiteboard.

Two proving grounds

The closing arc had two events in it, and they tested different claims. CAST put the method in front of testers and engineers — does the process hold up? BowFest put the product in front of archers on a course — does the thing actually work for the people it was built for?

BowFest didn't move the download counter. It redrew the roadmap instead — which, eighteen months in, is the more honest measure of whether any of this was real. The whole series that led here is in one place; the CAST side of the story is the next and final post.

Thanks for following the arc all the way to a 3D course.